Age exacerbates microglial activation, oxidative stress

RESEARCH Open Access

Age exacerbates microglial activation,oxidative stress, inflammatory and NOX2gene expression, and delays functionalrecovery in a middle-aged rodent model ofspinal cord injuryRamona E. von Leden1,2* , Guzal Khayrullina1,2, Kasey E. Moritz1 and Kimberly R. Byrnes1,2

Abstract

Background: Spinal cord injury (SCI) among people over age 40 has been steadily increasing since the 1980s and isassociated with worsened outcome than injuries in young people. Age-related increases in reactive oxygen species(ROS) are suggested to lead to chronic inflammation. The NADPH oxidase 2 (NOX2) enzyme is expressed by microgliaand is a primary source of ROS. This study aimed to determine the effect of age on inflammation, oxidative damage,NOX2 gene expression, and functional performance with and without SCI in young adult (3 months) and middle-aged(12 months) male rats.

Methods: Young adult and middle-aged rats were assessed in two groups—naïve and moderate contusion SCI.Functional recovery was determined by weekly assessment with the Basso, Beattie, and Breshnahan general motorscore (analyzed two-way ANOVA) and footprint analysis (analyzed by Chi-square analysis). Tissue was analyzed formarkers of oxidative damage (8-OHdG, Oxyblot, and 3-NT), microglial-related inflammation (Iba1), NOX2 component(p47PHOX, p22PHOX, and gp91PHOX), and inflammatory (CD86, CD206, TNFα, and NFκB) gene expression (all analyzed byunpaired Student’s t test).

Results: In both naïve and injured aged rats, compared to young rats, tissue analysis revealed significantincreases in 8-OHdG and Iba1, as well as inflammatory and NOX2 component gene expression. Further,injured aged rats showed greater lesion volume rostral and caudal to the injury epicenter. Finally, injuredaged rats showed significantly reduced Basso–Beattie–Bresnahan (BBB) scores and stride length after SCI.

Conclusions: These results show that middle-aged rats demonstrate increased microglial activation, oxidativestress, and inflammatory gene expression, which may be related to elevated NOX2 expression, and contributeto worsened functional outcome following injury. These findings are essential to elucidating the mechanismsof age-related differences in response to SCI and developing age-appropriate therapeutics.

Keywords: Spinal cord injury, Aging, NOX2, Microglia, Inflammation

* Correspondence: [email protected] Program, Uniformed Services University, 4301 Jones BridgeRoad, Bethesda, MD 20814, USA2Department of Anatomy, Physiology, and Genetics, Uniformed ServicesUniversity, Room C2099, 4301 Jones Bridge Road, Bethesda, MD 20814, USA

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

von Leden et al. Journal of Neuroinflammation (2017) 14:161 DOI 10.1186/s12974-017-0933-3

http://crossmark.crossref.org/dialog/?doi=10.1186/s12974-017-0933-3&domain=pdf

http://orcid.org/0000-0002-2554-6451

mailto:[email protected]

http://creativecommons.org/licenses/by/4.0/

http://creativecommons.org/publicdomain/zero/1.0/

BackgroundThere are over 250,000 people in the USA currentlyliving with a spinal cord injury (SCI), and over 15,000new cases are documented every year [1]. Theseinjuries cause permanent motor, autonomic, and sen-sory function loss [2]. Reduced cell viability andneuron, astrocyte, and oligodendrocyte loss are seenacutely after SCI [3]. White matter and axon loss,glial scar formation, and inflammation are seenchronically after injury, in the days to months follo-wing onset [3].SCI is largely studied in the young adult population

due to the high propensity of injuries from athletic andmilitary injuries and car accidents; however, there hasbeen a notable increase in age of onset in recent decades[4]. In the 1970s, the average age at onset of injury was29, but the average age has now increased to 42 [1].Aging tissue shows progressive DNA damage beginningby the age of 40, causing increased expression of genesinvolved in stress response and repair, including immuneand inflammatory genes [5]. Further, aging causeschanges to hormones and cellular function associatedwith increased co-morbidities, lengthened recovery time,reduced functional recovery, and prolonged chronicinflammation after SCI [6].One notable change to the cellular environment

with aging is an increase in oxidative stress caused bythe buildup of reactive oxygen species (ROS) overtime [7, 8]. The superoxide producing NADPH oxi-dase (NOX) enzyme is a major source of ROS [9].The NOX2 isoform has been shown to be the mostresponsive to injury and is present on microglia, theprimary inflammatory cell type of the central nervoussystem [10]. Microglia are found in a resting state inuninjured tissue, with ramified cell body and longprocesses. When activated, they take on a pro-inflammatory state with an amoeboid-like cell shape,releasing cytotoxic and inflammatory mediators [11].Activated microglia are found in the spinal cord cel-lular environment within 24 h post-injury and have amaintained, low-level activated presence at 60 dayspost-injury (dpi) [11].In animal models of traumatic brain injury (TBI), aged

microglia show alterations to morphology and activationcompared to younger microglia [12–14]. In uninjuredaged brain tissue, microglia demonstrate a hypertrophicor partially activated cell shape, considered a “primed”activation state [14]. When injured, these aged cellsshow a hyperactive response to injury compared toyounger cells, resulting in extended chronic inflamma-tion [14]. Previous work has demonstrated that in thespinal cord, compared to the brain, there is a signifi-cantly greater inflammatory response and recruitment ofmacrophages and neutrophils to trauma [15–17].

Therefore, findings in the aging brain may not accuratelyindicate what will occur in the spinal cord.Few studies have examined how age influences micro-

glial activation and/or oxidative stress in the spinal cord[18–21] and the role NOX2 plays in this process [22].Work in our lab has characterized the role of NOX2 inSCI [23], demonstrating that inhibition of NOX2 im-proves functional recovery and decreases oxidative stressand inflammation in a young adult rodent model of SCI[24]. Previous work on age-related cellular alterationsafter spinal cord injury have been performed in mice[22] and aged female rats [19], while in this paper weexamine a middle-aged male rat model of SCI as theaverage age of onset of human SCI is 42 years old. Basedon previous work showing age-related alterations afterinjury in the brain [12, 14, 25] and spinal cord [18, 19,21, 26], we investigated microglia, NOX2 components,and ROS prior to and chronically after injury(30 dpi). We now show that microglial and NOX2gene expression and ROS production are increased inmiddle-aged rats, demonstrating age-related alter-ations in the spinal cord.

MethodsAnimalsMale Sprague Dawley rats were given free access to foodand water and a 12-h light/12-h dark cycle. Young adultrats (3 months old: 300–500 g; Taconic, Germantown,MD: n = 24; Harlan, Frederick, MD: n = 6) from Harlanwere used in addition to rats from Taconic in pilot naïveand 30 dpi injured study groups for immunohistochem-istry and functional assessment. No significant differencewas noted between groups from different vendors;therefore, data from all vendors are presented together.Rats were housed for 1 week prior to undergoing sur-gery or assessment for naïve studies. Middle-aged rats(12-months-old, retired breeders: 400–600 g; Harlan,Frederick, MD: n = 28) were acquired at 10 monthsof age and gentled twice a week until they had ma-tured to 12 months of age, at which point theyunderwent surgery or assessment for naïve studies.One 12-month-old injured rat naturally perished at20 dpi and was removed from behavioral testing data(final n = 17 behavioral analysis). Two 3-month-oldinjured rats were not included in tissue analysis astissue was damaged in processing (final n = 14 tissueanalysis). All 12-month-old injured rats were sub-jected to full behavioral battery, but the behavioralbattery for the first cohort of 3-month-old injuredrats did not include footprint analysis testing, whichmeans that the total number of animals tested in the3-month-old group was smaller than that in the 12-month-old rats (n = 5/10 3-month-old rats able toperform 28 dpi footprint analysis, n = 7/18 12-

von Leden et al. Journal of Neuroinflammation (2017) 14:161 Page 2 of 14

month-old rats able to perform 28 dpi footprint ana-lysis). After functional testing, groups were dividedamong tissue analysis groups for histological, immu-nohistochemical, or biochemical analysis as designatedprior to injury. All animal procedures were approvedby the Uniformed Services University IACUC andcomplied fully with the principles set forth in the“Guide for the Care and Use of Laboratory Animals”prepared by the Committee on Care and Use ofLaboratory Animals of the Institute of Laboratory Re-sources, National Research Council (DHEW pub. No.(NIH) 85-23, 2985); this manuscript has been writtenin accordance with the ARRIVE (Animal Research:Reporting In Vivo Experiments) guidelines [27].

Contusion spinal cord injuryContusion SCI was performed in rats (n = 35) (see Table 1for breakdown of animal groups) as described previously[28]. Briefly, rats were anesthetized with isoflurane (4%induction, 2.5% maintenance in 50% O2), and a laminec-tomy was performed at vertebral level T9 to expose thespinal cord. Moderate injury (150 kDynes force, 1500-μdisplacement, 127-mm/s velocity, 1-s dwell time) wasinduced with the Infinite Horizons Impactor (PrecisionSystems Incorporated, Natick, MA). Naïve animals(n = 22) received no surgery or anesthesia. Post-injurycare included dual housing on diamond soft bedding(Harlan Laboratories, Frederick, MD), twice daily bladderexpressions until spontaneous micturition returned, andanalgesic supplementation to drinking water for 48 hfollowing injury (Children’s Tylenol; acetaminophen,200 mg/kg).

Functional assessmentThe Basso–Beattie–Bresnahan (BBB) scale was used toassess functional neurological recovery, as detailed previ-ously [29]. Performance of the left and right hind limbson a scale from 0 (no spontaneous movement) to21(normal locomotion) was averaged to obtain the totalBBB on 1, 7, 14, 21, and 28 dpi. More detailed locomo-tion was assessed using the “footprint analysis test,”modified from Kunkel-Bagden et al. [30] and performedon the same days as BBB assessment. Subjects’ hind

paws were pressed to an inkpad to dye their hind feetand were then allowed to walk across a sheet of paperlining a long and narrow runway (6.5 in. wide, 76 in.long) to ensure animals walked in a straight line for ana-lysis. Prints were analyzed for width of toe spread andlength of stride length (in mm). Subjects’ footprints wereonly analyzed if they had regained ability to walk withsupport on the hind limbs (a minimum BBB score of11), and percentage of animals who had regained theability to perform the task was analyzed by age group ateach assessment day.

Histology and lesion volumeAt 30 dpi, rats were anesthetized with Euthasol (pento-barbital sodium and phenytoin sodium; 200 mg/kg, IP)and intracardially perfused with 10% buffered formalin(Fisher Scientific, Hampton, New Hampshire). A 1-cmblock of spinal cord tissue was dissected, extending5 mm rostral and caudal from the center of the lesionsite (or from the T9 vertebral site for naïve animals).Tissue was sectioned into 20-μm axial sections, andevery third slice was saved for histological analyses. Todetermine lesion volume, tissue was processed with astandard H&E stain. Lesion volume from H&E imagescollected with NanoBrightfield Software (HamamatsuCorporation, Middlesex, NJ) were quantified using theCavalieri method as previously described [31, 32]. Tensections were analyzed per animal at even intervalsthroughout the spinal cord from 360-μm pre-lesionepicenter (0–360 μm on histogram in Fig. 5a) throughlesion epicenter (between 360 and 540 μm on histogram)to 360 μm post-epicenter (540–900 μm on histogram) toprovide an analysis of volume of lesion throughout thespinal cord.

ImmunohistochemistryStandard fluorescent immunohistochemistry was per-formed on tissue sections from animals 30 dpi obtainedas described above. Iba1 (microglia, 1:100; Wako Biopro-ducts, Cat# 019-19741 RRID:AB_839504; Richmond,VA) and 8-OHdG (DNA oxidation, 1:500; Abcam, Cat#ab62623 RRID:AB_940049; Cambridge, MA) were usedas primary antibodies, with visualization by fluorescent

Table 1 Number of animals per group for all experiments in study

Number of animals per group and experiment

Age Injured/naïve Function Immunohistochemistry Lesion volume Biochemistry/PCR

3 months old Naïve 12 4 0 4

12 months old Naïve 10 4 0 4

3 months old Injured 18 4 6 4

12 months old Injured 17 4 6 4

Animals were first divided by age (3 or 12-months old) and injury type (naïve or injured). Functional analysis was performed on all animals in study. Animals usedin function experiments were sacrificed and divided into subgroups for tissue experiments (immunohistochemistry, lesion volume, and biochemistry/PCR)


secondary antibodies (Alexa-Fluor Secondaries, 1:500;Molecular Probes, Eugene, OR). Negative controls (sectionsnot stained with primary antibody) were used to confirmspecificity of secondary antibodies. Immunofluorescent im-ages were captured under the same fluorescence conditionsand exposure time on a NanoZoomer system (Hamamatsu,Bridgewater, NJ) or an Olympus BX43 microscope (Olym-pus America). Immunofluorescence was quantified usingpixel density measurement in Scion Image as previously de-scribed [33]. Ten axial sections were analyzed per animalstarting caudally at the first sign of lesion, then at even in-tervals rostrally through the spinal cord for 1 cm to provideanalysis throughout the lesion. Analysis was performed ongray matter directly adjacent to the lesion site at 20×magnification. The average pixel density of all ten sectionswas reported for each animal.

Comparative RT-PCRA 10-mm section of the spinal cord, centered at the le-sion epicenter, was dissected and immediately frozen ondry ice. The TRIzol® protein and mRNA extractionmethod (Invitrogen, Carlsbad, CA) was employed so thateach sample could be used for both western blotting andmRNA quantification. Briefly, RNA was extracted fromeach cord sample individually using TRIzol® reagent andchloroform, then purified with the Qiagen RNeasy kit(#74104, Qiagen, Hilden, Germany). RNA concentrationwas measured using the Nanodrop system (Thermo Sci-entific, Wilmington, DE). Complementary DNA (cDNA)conversion was performed from 1 μg of RNA using aVeriti thermal cycler (Applied Biosciences, Waltham,MA) and a high-capacity cDNA conversion kit (AppliedBiosciences, Waltham, MA) using the manufacturer’sprotocol. Quantitative real-time PCR (qRT-PCR) was per-formed using the StepOnePlus Real-Time PCR System

(Applied Biosciences, Waltham, MA) according to themanufacturer’s instructions. To determine changes in geneexpression, primers were designed using the Primer-Blasttool, then obtained from Integrated DNA Technologies(IDT, Coralville, IA; see Table 2 for detailed primer infor-mation). Reported gene expression values were determinedusing the ΔΔ CT method: raw value of each samplenormalized first to internal control (GAPDH) and then tothe control group (3 months).

Western blot and OxyblotProtein from frozen tissue samples was isolated in TRIzol®reagent (Invitrogen, Carlsbad, CA). Detection of proteinoxidation was performed by processing protein samplesusing the Oxyblot Protein Oxidation Detection AnalysisKit (Millipore, Billerica, MA) according to the manufac-turer’s protocol. For both Oxyblot and western blotting,15 μg of sample protein was run in a Mini-PROTEAN®TGX™ Precast Gel (Bio-Rad, Hercules, CA) and trans-ferred to a Trans-Blot® Turbo™ (Bio-Rad) nitrocellulosemembrane. Antibodies raised against 3-nitrotyrosine(5 μg/ml; Abcam, Cat# ab61392, RRID:AB_942087) wereprobed overnight at 4 °C. Immune complexes and oxi-dated proteins were detected with appropriate secondaryantibodies and chemiluminescence reagents (Pierce).For both Oxyblot and western blots, GAPDH (0.5 μg/

ml; Millipore, Cat# MAB374 RRID:AB_2107445) wasused as a control for gel loading and protein transfer.NIH ImageJ was used to assess pixel density of resultantblots for quantitation.

StatisticsSample sizes were calculated based on our previouswork [34]. All assays were carried out by investigatorsblinded to subject group. Quantitative data are presented

Table 2 Gene primers tested in comparative RT-PCR experiments

Gene Sense Antisense

CD86 5′-TGCTCATCTAAGCAAGGATACCCG-3′ 5′-CGACTCGTCAACACCACTGTCCTG-3′

CD206 5′-GGATTGTGGAGCAGATGGAAG-3′ 5′-CTTGAATGGAAATGCACAGAC-3′

p22PHOX 5′-GACGCTTCACGCAGTGGTACT-3′ 5′-CACGACCTCATCTGTCACTGG-3′

p47PHOX 5′-CAGAATGTTGCCTGGTTG-3′ 5′-GTCCCCTCCCTTAGATGA-3′

gp91PHOX 5′-CTTCACACGGCCATTCACAC-3′ 5′-GTCATAGGAGGGTTTCCGGC-3′

TNFα 5′-GGCAGCCTTGTCCCTTGAAGAG-3′ 5′-GTAGCCCACGTCGTAGCAAACC-3′

NFκB 5′-GTGCAGAAAGAAGACATTGAGGTG-3′ 5′-AGGCTAGGGTCAGCGTATGG-3′

iNOS 5′-CAGCCCTCAGAGTACAACGAT-3′ 5′-CAGCAGGCACACGCAATGAT-3′

Arg-1 5′-CAGTATTCACCCCGGCTACG-3′ 5′-GCCTGGTTCTGTTCGGTTTG-3′

CD68 5′-TGTACCTGACCCAGGGTGGAA-3′ 5′-GAATCCAAAGGTAAGCTGTCCGTAA-3′

IL10 5′-CAAGGAGCATTTGAATTCCC-3′ 5′-GGCCTTGTAGACACCTTGGTC-3′

GAPDH 5′-GCTGGTCATCAACGGGAAA-3′ 5′-ACGCCAGTAGACTCCACGACA-3′

Table provides sense and antisense sequences for each gene tested in this study. Genes investigated that showed no significant difference by age in either naïveor injured animals were omitted from results


as mean ± standard error of the mean (SEM). Generalmotor performance using the BBB test was analyzedusing a two-way ANOVA with Tukey’s multiple compa-rison’s post hoc test. Ability to plantar step in footprintanalysis task was analyzed using Chi-square analysis.Footprint analysis test, immunohistochemistry, biochem-istry, and qRT PCR were analyzed using the Student ttest. All statistical tests were performed using theGraphPad Prism Program, Version 6.03 for Windows(GraphPad Software, San Diego, CA). A p value < 0.05was considered statistically significant.

ResultsUninjured middle-aged animals show altered steppingpattern in footprint analysisFootprint analysis of toe spread and stride length inuninjured animals revealed an alteration in motorfunction in middle-aged animals compared to younganimals (Fig. 1a). Toe spread was measured from thethumb toe (toe 1) to the pinky toe (toe 5) in the hindpaws. Twelve-month-aged rats showed a significantlywider toe spread than 3-month-aged rats (t(20) = 4.98,p = 0.0005). Stride length was measured as the length inmillimeter between each consecutive step in the hindpaws. Though not significant, 12-month-aged ratsshowed a trend toward longer strides than 3-month-aged rats (Fig. 1b; t(20) = 1.97, p = 0.0627), indicatingthat overall stepping patterns are altered with age.

Microglial activation and pro-inflammatory gene expres-sion is altered with ageTo determine if aged tissue showed basal differences inmicroglial activation, uninjured tissue from 3- and 12-month-old rats was immunostained with the microglial

marker Iba-1 (Fig. 2a) and processed for comparativeRT-PCR for expression genes associated with microglialactivation and NOX2 components. Our findings showedsignificantly increased Iba-1 staining (t(6) = 2.59,p = 0.0496) in 12-month-aged rat tissue compared to 3-month-aged rats (Fig. 2b). Comparative RT-PCR results(Fig. 2c) showed significantly increased gene expressionin 12-month-aged rats of the pro-inflammatory pheno-type marker CD86 (t(6) = 7.46, p = 0.0050) compared to3-month-aged rats. No significant difference by age wasfound in any other genes probed, including the anti-inflammatory phenotype marker CD206, transcriptionfactor NFκB, or inflammatory cytokine TNFα (Table 3).Relative fold changes between 3- and 12-month-old ratsfor all genes tested listed in numerical form can befound in Table 3 under “naïve.”

Middle-aged rats show increased gene expression ofNOX2 cytosolic subunit component p47PHOX

To determine if NOX2 activity and expression were influ-enced by age, uninjured tissue from 3- and 12-month-oldrats was processed for comparative RT-PCR and analyzedfor target gene expression. Gene expression of theactivated cytosolic subunit p47PHOX was significantlyincreased in 12-month-aged rats compared to that in 3-month-aged rats (t(6) = 3.47, p < 0.0001; Fig. 2d). How-ever, gene expression of the activated membrane boundNOX2 components p22PHOX and gp91PHOX did not revealany significant difference by age.

ROS production is altered with ageTo determine if tissue showed basal differences in ROSproduction with age, uninjured tissue from 3- and 12-month-old rats was immunostained with the DNA

a bToe Spread

mm

3 months 12 months19

20

21

22

***

Stride Length

mm


75

80

85

90

Fig. 1 Uninjured aged animals show altered stepping pattern in footprint analysis. a Uninjured 12-month-old rats (n = 10) have significantly widertoe spread than 3-month-old rats (n = 12). b Uninjured 12-month old-rats show a trend toward increased stride length compared to 3-month-oldrats, supporting toe spread findings. ***p < 0.001. Bars represent mean ± SEM


oxidation marker 8-OHdG (Fig. 3a), Oxyblot analysiswas performed to determine protein oxidation, andwestern blotting was performed to determine levels of 3-NT, a marker of protein nitrosylation. Our findingsshowed that 12-month-aged rats compared to 3-month-aged rats demonstrated a significant increase in 8-OHdG staining (t(6) = 3.06, p = 0.0281; Fig. 3b) and atrend toward increased protein oxidation via Oxyblot(t(6) = 2.18, p = 0.0946; Fig. 3c). Western blotting didnot show any difference in protein nitrosylation betweenthe two ages (Fig. 3d).

Middle-aged rats show diminished motor functionrecovery after spinal cord injuryNext, we examined both middle-aged and young animalsafter moderate contusion SCI to investigate how age

alters response to injury. Our results from the BBBgeneral motor test showed worsened performance byaged animals after injury (Fig. 4). Twelve-month-old ratsshowed worsened performance compared to 3-month-old rats at all time points except 1 dpi, indicating signifi-cantly impaired recovery to injury [F(1.33) = 12.81,p < 0.0001; Fig. 4a]. At 1 dpi, 12- and 3-month-old ratsaveraged a score of 0.471 ± 0.2827 and 1.000 ± 0.485 re-spectively, indicating no stepping ability with only slightmovement of one joint on one limb. At 7 dpi, the 3-month-old group has progressed to a mean of6.583 ± 1.188, indicating extensive movement in all mul-tiple joints, while the 12-month-old group remained at anaverage score of 0.529 ± 0.208. By 14 dpi, the 3-month-oldgroup continued to show steady improvement with a BBBscore of 10.970 ± 1.217, with occasional weight-supported

c

1.25x

Iba1

3 months 12 months

20x 20x

1.25x

CD86ge

ne e

xpre

ssio

n

3 months 12 months0

1

2

3

6 10 06

8 10 06

1 10 07

**

CD206

gen

e ex

pre

s sio

n

3 months 12 months0.0

0.5

1.0

1.5

2.0

2.5

gp91PHOX

gen

e ex

pre

ssio

n


0.5

1.0

1.5

2.0

2.5

p47 PHOX

gen

e ex

pre

ssio

n

3 months 12 months0

5

10

15

20

25

30000

40000

50000

****

Iba1

Pix

els/

mm

2

3 months 12 months0

50000

100000

150000

*

p22 PHOX

gen

e ex

pre

ssio

n

3 months 12 months0

1

2

3

4

5

b

d

a

Fig. 2 Microglia and NOX2 component activation and gene expression is altered with age. a Fluorescent microscope images of longitudinal sectionsof naïve 3- and 12-month-old rat spinal cords. Sections were stained with Iba-1 (green). Sections were imaged at 1.25× magnification and encompassan entire 10-mm section in the thoracic spinal cord. Inserts were taken at 20× magnification in the gray matter delineated by white boxes and arrows. bImmunostaining shows age-related differences in the microglial marker Iba1, showing significantly increased staining in 12- versus 3-month-old rats(size bar = 100 um). N = 4/group. c Comparative qPCR reveals significant increase in gene expression of the pro-inflammatory phenotype markerCD86. d Comparative RT-PCR reveals significant increase in p47PHOX gene expression in 12-month-aged rats compared to 3-month-aged rats. Nosignificant difference was seen in gene expression of p22PHOX or gp91PHOX between the two groups. N = 4/group. *p < 0.05, **p < 0.005,****p < 0.0005. Bars represent mean ± SEM


steps but no limb coordination, and the 12-month-oldgroup showed sweeping of the feet with no weight-bearing steps, averaging 6.971 ± 1.309. At 21 and28 dpi, improvement in the 3-month-old group wasstill steady (13.310 ± 1.009 and 14.42 ± 1.070, re-spectively), showing consistent weight-bearing plantarstepping with coordinated limb movements, while the12-month-old group reached a max performance of9.618 ± 1.202 at 21 dpi, with weight-bearing plantarstepping but only occasional coordinated limb move-ments and never improved beyond that score. Thiscomparison demonstrates not only that the 12-month-old groups’ recovery was diminished in comparison withthat of the 3-month-old but also that the overall speed ofrecovery was delayed—the improvements in functionalperformance in the 12-month-old group occurred at least1 week after the same score had been achieved in the 3-month-old group.The number of rats that achieved plantar stepping, en-

abling them to complete footprint analysis, at each timepoint through 28 days was assessed and found to differby age. The number of 12-month-aged rats that achievedplantar stepping by 28 dpi was significantly lower thanthe 3-month-aged rats [c2(3, N = 31) = 37.52,p < 0.0001; Fig. 4b]. Rats that recovered the ability totake plantar steps with their hind limbs were tested onthe footprint analysis task. At 28 dpi, 12-month-old ratsthat were able to plantar step demonstrated a signifi-cantly shorter stride length (t(10) = 2.53, p = 0.0297;Fig. 4d) and a trend toward narrower toe spread(t(10) = 1.192, p = 0.0967; Fig. 4c) in comparison tothe 3-month-old rats.

Middle-aged rats have significantly greater lesion volumerostral and caudal to the lesion epicenterAt 30 dpi, results from our lesion volume analysisshowed a larger lesion volume in 12-month-old ratscompared to that in 3-month-old rats; however, the vol-ume of the lesion varied by proximity to lesion epicenter(Fig. 5a). At the lesion epicenter (360–540 μm), both 3-and 12-month-old animals demonstrated marked tissuedamage and loss, with no significant difference in vol-ume between age groups (Fig. 5c). But in both rostral(0–180 μm) and caudal (720–900 μm) lesion epicenters,12-month-old animals had significantly greater lesionvolumes than 3-month-old animals (t(10) = 6.09,p < 0.0001/rostral, and t(10) = 4.64, p = 0.0017/caudal;Fig. 5b and d).

Microglial activation and gene expression 30 days afterspinal cord injury is altered with ageImmunostaining at 30 dpi in both 3- and 12-month-aged tissues revealed increased microglial activation, asdefined by cell size and number, and reflected in pixeldensity of Iba1 staining microglia in comparison to naïvetissue (Fig. 6a). Aged tissue showed significantly greaterIba1 staining compared to young tissue (t(6) = 2.16,p = 0.0487; Fig. 6b). Comparative RT-PCR results (Fig. 6c)showed significantly increased gene expression in 12-month-aged rat compared to 3-month-aged rats ofpro-inflammatory marker CD86 (t(6) = 7.26,p = 0.0050), anti-inflammatory marker CD206(t(6) = 5.75, p = 0.0045), inflammatory cytokine TNFα(t(6) = 3.47, p = 0.0127), and transcription factorNFκB (t(6) = 4.98, p = 0.0156). Relative fold changes

Table 3 Comparative gene fold changes for all genes tested by injury group

Comparative fold change by injury group

Naïve Genetested

30 dpi

3 months old 12 months old 3 months old 12 months old

1.467 ± 0.8292 **6.993 × 106 ± 1.258 × 106 CD86 1.096 ± 0.2886 ***1.253 × 106 ± 145,818

1.371 ± 0.5850 0.7830 ± 0.2167 CD206 1.104 ± 0.3465 **139,991 ± 24,361

2.217 ± 1.785 0.6052 ± 0.04079 p22PHOX 1.030 ± 0.1825 *20678 ± 6017

8.352 ± 8.037 ****39,131 ± 2480 p47PHOX 7.739 ± 5.672 *1.225 × 106 ± 473,918

0.9453 ± 0.05264 1.583 ± 0.4531 gp91PHOX 0.9421 ± 0.08263 *1.948 ± 0.4574

3.229 ± 2.839 163.1 ± 141.8 TNFα 0.6954 ± 0.3801 *11.55 ± 2.502

1.226 ± 0.5548 0.8387 ± 0.6209 NFκB 1.027 ± 0.1751 *4.108 ± 0.7734

44.06 ± 22.85 0.01669 ± 0.01419 iNOS 1.109 ± 0.3358 1.069 ± 0.4000

3.118 ± 2.773 1230 ± 794.0 Arg-1 1.047 ± 0.2304 1.382 ± 0.03944

0.4958 ± 0.4531 0.3324 ± 0.02472 CD68 1.104 ± 0.3346 1.011 ± 0.3394

2.438 ± 1.521 7.806 ± 4.643 IL10 1.051 ± 0.2427 1.057 ± 0.1975

Table demonstrates how age affects gene expression before and after injury. The average of the 3-month-old group in both the naïve and injury groups wasconsidered the control mean value. Each individual sample’s fold change was compared to the control mean. Fold changes are presented in table as mean valuesof all samples in each group ± SEM. N = 4/group*p < 0.05, **p < 0.005, ***p < 0.0005


between 3- and 12-month-old rats for all genes testedlisted in numerical form can be found in Table 3under “30 dpi”. An additional experiment comparinggene expression between naïve and injured groupswithin each age group was performed on selectedgenes to verify gene expression patterns (Table 4).

Middle-aged rats show increased NOX2 component geneexpression at 30 dpiTo determine if NOX2 activity and expression after in-jury were influenced by age, tissue from injured 12- and3-month-aged rats at 30 dpi was processed for compara-tive RT-PCR and analyzed for expression of target genes.Gene expression of the cytosolic subunit p47PHOX andthe membrane components p22PHOX and gp91PHOX

were significantly increased in 12-month-aged rats

compared to 3-month-aged rats (t(6) = 3.47, p = 0.0404;t(6) = 4.61, p = 0.0182 and t(6) = 2.94; p = 0.0186,respectively; Fig. 6d).

Middle-aged rats show increased ROS production andoxidative stress at 30 dpiTo determine if injured aged tissue showed differencesin ROS production, tissue from 3- and 12-month-oldrats was processed for Oxyblot and western blotting at30 dpi. Oxyblot analysis revealed that 12-month-agedrats when compared to 3-month-aged rats demonstratedsignificantly increased protein oxidation (t(6) = 4.04,p = 0.0156; Fig. 7a), and western blotting of 3-NT, amarker of protein nitrosylation, demonstrated signifi-cantly increased protein nitrosylation (t(6) = 1.758,p = 0.0038; Fig. 7b).

b c d 3-NT

3-N

T/G

AP

DH


0.2

0.4

0.6

0.8

1.0Oxyblot

no

rmal

ized

to

co

ntr

ol a

nd

gap

dh

3 months 12 months0

2

4

6

8

10

8-OHdG

Pix

els/

mm

22

3 months 12 months0

1000

2000

3000

4000

5000

*

3 m 3 m12 m 12 m

GAPDH

3-NT

12 mSample

12 mControl

3 mSample

3 mControl

Gapdh

20x

1.25x1.25x

12 months3 months

8-OHdG

20x

a

Fig. 3 ROS production is increased with age. a Fluorescent microscope images of longitudinal sections of naïve 3- and 12-month-old rat spinal cords.Sections were stained with 8-OHdG (green). Sections were imaged at 1.25× magnification and encompass an entire 10-mm long section in the thoracicspinal cord. Inserts were taken at 20× magnification in the gray matter delineated by white boxes and arrows. b 8-OHdG stain shows significantly greateroxidative damage in 12 versus 3-month-old rats (size bar = 300 um). N = 4/group. c Oxyblot analysis shows a trend (p = 0.0946) toward increased proteinoxidation in 12-month-aged rats compared to 3-months-aged rats. d Western blot analysis that shows protein expression of 3-NT, a marker of nitrosylatedproteins which is a product of oxidative stress, is not significantly different between 12- and 3-month-aged rats. N = 4/group. *p < 0.05. Barsrepresent mean ± SEM


DiscussionThis study demonstrates that aging alters the inflamma-tory environment in the 12-month-old rat spinal cord.Uninjured 12-month-old, middle-aged rats showincreased microglial activation and NOX2 cytosolicsubunit and pro-inflammatory gene expression and in-creased DNA oxidation when compared to 3-month-old,young adult rats. At 30 dpi, middle-aged rats show de-creased functional recovery and increased lesion volume,microglial activation, NOX2 component and inflamma-tory gene expression, and protein oxidation and nitrosy-lation compared to young adult rats.

Previous work has shown increasing oxidative stresswith age [8]. We now demonstrate that this increase isobvious in the CNS, with increased markers of oxidativestress in the middle-aged rat spinal cord both before andafter injury. In naïve middle-aged rats compared toyoung adult, increases in the DNA peroxidation marker8-OHdG, but not of protein carbonylation (Oxyblot) andnitrosylation (3-NT), were observed. As mutations inDNA have been found by early middle age [5], DNAperoxidation may be one of the first signs of increasingROS production, as part of a pro-inflammatory environ-ment in an aging tissue [14]. As protein modifications

a b

c d

BBB

DPI

BB

B S

core

0 10 20 30

0

5

10

15

20

25

3 Months 12 Months

****

** ***

Naive

Stride Length

mm


70

80

90

100

*

Toe Spread

mm


17

18

19

20

Successful Plantar Stepping

Per

cen

t

3 months 12 months0

50

100

150

7 dpi 14 dpi 21 dpiDid not achieve

* * * *

Fig. 4 Aged animals show diminished motor function recovery after injury. a At 1 dpi, no difference in performance in BBB scores were observed ininjured rats. At all other time points (7, 14, 21, and 28 dpi), 12-month-old injured rats (n = 17) showed significantly reduced BBB scores compared to 3-month-old rats (n = 18), indicating significantly impaired recovery to injury. b Aged animals show impaired ability to plantar step and complete footprintanalysis task. Twelve-month-old rats (n = 17) show significantly diminished ability to plantar step in order to complete the footprint analysis task comparedto 3-month-old rats (n = 14), indicating a diminished ability to take steps or strides. c At 28 dpi, toe spread analysis showed a trend in 12-month-old rats(n = 7) toward narrower toe spread compared to 3-month-old rats (n = 5). d 12-month-old rats (n = 7) showed significantly decreased stride lengthcompared to 3-month-old rats (n = 5). *p < 0.05, **p < 0.005, ***p < 0.0005, ****p < 0.0001. Bars represent mean ± SEM


occur further downstream than DNA, age-related in-creases in protein carbonylation and nitrosylationmarkers may be more indicative of a hyperactive inflam-matory response to injury.The increased ROS production found in middle-aged

rats coincides with our results showing increasedmicroglial activity with age. Increased staining of themicroglial marker Iba1 was coupled with increased pro-inflammatory gene expression of CD86 in middle-agedrats, suggesting that microglia are present in aged tissuein a pro-inflammatory state. Further, staining demon-strates alterations in microglial phenotype as definedpreviously [11], with 3-month-old rats showing a non-activated phenotype with lengthy processes and a smallcell body and 12-month-old rats showing a partially

activated microglial phenotype, with an engorged,rounder shaped cell body. These findings in the spinalcord are similar to previous findings in the brain [14],suggesting an age-related pattern of the inflammatoryresponse that is not impacted by regional differences be-tween the spinal cord and the brain [17].Interestingly, our gene expression experiments at 30 dpi

showed an increase in both pro-inflammatory- and anti-inflammatory-related genes, suggesting that chronicallyafter injury, all microglial activities are elevated, not onlypro-inflammatory as seen in uninjured tissue. Previouswork using aged mouse and female rat models of trau-matic injury has shown a decrease in anti-inflammatorysignaling and an increase in pro-inflammatory signaling at24 h and 7 dpi [18, 19, 21, 25]. In a traumatic brain injury

12 months3 months 12 months3 months3 months 12 months

a

b dcLesion Volume

Rostral to Epicenter

mm

3

0

5

10

15

****

Lesion Volume

Epicenter

mm

3

3 months3 months 12 months12 months0

10

20

30

40

Lesion Volume

Caudal to Epicenter

mm

3

3 months 12 months0

5

10

15

20

**

Lesion Volume Histogram

Distance inmicrons

mm

3

0 200 400 600 800 1000

0

20

40

60

3 months

12 months

e

Fig. 5 Aged rats have significantly great lesion volume rostral and caudal to the lesion. a Quantitative analysis of tissue stained with H&E reveals thatat 30 dpi, 12-month-old rats show an increased lesion volume compared to 3-month-old rats both rostral and caudal to the injury epicenter, suggest-ing that aged rats have an increased volume and length of lesion compared to young. Lesion volume histogram shows the pattern of lesion spreadfrom the epicenter. b Lesion volume rostral to the epicenter (0–180 μm) was significantly greater in 12-month-old rats compared to 3-month-old rats.c No significant difference was found in lesion volume at injury epicenter (360–540 μm). d Lesion volume caudal to the epicenter (720–900 μm) wassignificantly greater in 12-month-old aged rats compared to 3-month-old rats. e Photos depict sections of the spinal cord at each segment throughthe lesion from both 3- and 12-month-old rats. Rostral image is from a 120-μm section, epicenter image is from a 400-μm section, and caudal image isfrom a 820-μm section. Size bar = 3 mm. N = 6/group. **p < 0.005, ****p < 0.0001. Bars represent mean ± SEM


a

c

dp22PHOX

gen

e ex

pre

ssio

n

0

1

2

10000

20000

30000 *p47 PHOX

ge

ne

ex

pre

ss

ion

3 months 12 months0

10

20

500000

1000000

1500000

2000000*

CD86

gen

e ex

pre

ssio

n

3 months 12 months0

1

2

1000000

1500000

2000000

***

CD206

gen

e ex

pre

ssio

n

3 months 12 months

0

1

2100000

150000

200000

**

NF B

gen

e ex

pre

ssio

n

3 months 12 months0

2

4

6

*

Iba1

Pix

els/

mm

2

3 months 12 months0

100000

200000

300000

400000

*


0.5

1.0

1.5

2.0

10

15 *

TNF

gen

e ex

pre

ssio

n

gp91PHOX

gen

e ex

pre

ssio

n

0

1

2

3

*

12 months3 monthsIba1

H&E 2.5x2.5x

20x 20x

b

3 months 12 months 3 months 12 months

Fig. 6 Microglial and NOX2 component activation and gene expression after spinal cord injury at 30 dpi. a 30 dpi H&E-stained images of 3 and12-month-old injured spinal cord. Black boxes depict location of Iba1-stained quantification, in the closest intact tissue outside the central cavityof the injury. Black arrows point to corresponding images of immunofluorescent staining of Iba1. b Twelve-month-aged rats showed significantlygreater staining in comparison to 3-month-aged rats at 30 dpi. N = 4/group. c Comparative RT-PCR reveals significant increase in gene expressionof the M1 phenotype marker cd86, M2 phenotype marker cd206, and inflammatory cytokines TNFα and NFκB in 12-month-aged rats comparedto that in 3-month-aged rats, suggesting an overall increase in microglial gene expression after injury with age. d Comparative RT-PCR of NOX2components p47PHOX, p22PHOX, and gp91PHOX reveal a significant increase in gene expression of all three genes in 12-month-aged rats comparedto 3-month-aged rats at 30 dpi. N = 4/group. *p < 0.05, **p < 0.005, ***p < 0.0005. Bars represent mean ± SEM

Table 4 Comparative gene fold changes by age group for significantly impacted genes

Comparative fold change by age group

3 months old Genetested

12 months old

Naïve 30 dpi Naïve 30 dpi

1.467 ± 0.8292 4.481 ± 1.179 CD86 0.6781 ± 0.3546 0.7444 ± 0.08664

1.731 ± 0.5850 0.7586 ± 0.5561 CD206 1.094 ± 0.3027 111,858 ± 67,300

1.043 ± 0.2966 225.2 ± 76.85 p22PHOX 1.005 ± 0.06771 218.9 ± 132.1

1.491 ± 0.8159 3.134 ± 2.551 p47PHOX 1.232 ± 0.4360 *180,416 ± 31,468

1.001 ± 0.02864 28.02 ± 3.662 gp91PHOX 1.154 ± 0.3588 *54.89 ± 21.85

Table demonstrates the effect of injury on gene expression within age groups. The average of the naïve in both the 3- and 12-month-age groups was consideredthe control mean value. Each individual sample’s fold change was compared to the control mean. Fold changes are presented in the table as mean values of allsamples in each group ± SEM. N = 4/group*p < 0.005


model with 23-month-old mice, a similar increase in bothpro- and anti-inflammatory gene expression was observedat 24-h post-injury [35].Our findings at 30 dpi suggestthat a change in activation type may occur chronicallypost-injury that leads to a more active overall inflam-matory response. Further, Morganti et al. [35] foundthat acute gene expression alterations were stronglydependent on peripheral monocyte infiltration; it ispossible that the 30 dpi data in our study is moremicroglial related due to the reduction in macrophageinvasion at this later time point [36]. Future workshould evaluate these earlier time points and contri-bution of peripheral and central inflammatory cells in12-month-old rats to fully characterize our rodentmodel.Increased gene expression of the activated NOX2 cyto-

solic subunit p47PHOX but not the membrane compo-nents p22PHOX or gp91PHOX suggests that NOX2expression is increased in middle-aged naïve rats, corre-sponding with the observed pro-inflammatory microglia.At 30 dpi, gene expression of both NOX2 membranecomponents p22PHOX and gp91PHOX and the cytosolic

subunit p47PHOX are elevated in middle-aged rats com-pared to young adults, suggesting that microglial NOX2is “primed” prior to injury and activated more fully withinjury. This could play a role in the increased ROSproduction we see in a middle-aged spinal cord. We andthe others have shown that NOX2 is involved in micro-glial ROS production [22–24].Our lesion volume experiments at 30 dpi showed that

at the epicenter of the injury, the lesion volume was notsignificantly different by age, but was significantly in-creased in middle-aged rats compared to young adults insections of the spinal cord both rostral and caudal to theinjury, demonstrating an increased volume farther fromthe lesion epicenter in middle-aged rats. Our resultscoincide with previous work showing increased lesionvolume in aged female rats after SCI [19] and in a mousemodel of TBI [25]. These findings demonstrate that in-jury severity is increased in middle-aged rats, which maybe a result of the exaggerated inflammatory responseleading to increased oxidative stress on cells. Control ofmotor function is distributed throughout the spinal cordwith multiple interacting tracts, and recovery of function

OxyblotN

orm

aliz

ed t

o

Co

ntr

ol a

nd

GA

PD

H

3 months 12 months 3 months 12 months0

2

4

6

8

*

3-NT

3-N

T/G

AP

DH

8

10

12

14**

3 m 12 m 12 m 3 m

3 mSample

3 mControl

12 mControl

12 mSample

GAPDH

GAPDH

3-NT

a b

Fig. 7 Aged rats show increased ROS production and oxidative stress at 30 dpi. a Oxyblot analysis reveals a significant increase in proteinoxidation in 12-month-aged rats compared to 3-month-aged rats at 30 dpi. b Western blot analysis reveals a significant increase in 3-NT, a markerof nitrosylated proteins which is a product of oxidative stress, in 12-month-aged rats compared to that in 3-month-aged rats 30 dpi. N = 4/group.*p < 0.05, **p < 0.005. Bars represent mean ± SEM


observable in injury models is largely due to the abilityof the spinal networks to compensate for some loss oftissue [37]. Interestingly, previous work has shown thatthe increase of lesion volume and functional deficits isnot completely linear, but with increasing injury severity,once a threshold proportion of tract is lost, loss of func-tion is seen [38, 39]. The increased lesion volume foundin our middle-aged rats would impact a greater numberof tracts in the spinal cord, likely surpassing that thresh-old and contributing to the delayed functional recoverywe observed.Interestingly, naïve middle-aged rats also show func-

tional impairments, with altered stepping patterns withwider toe spread and longer stride length compared toyoung adult rats. Functional performance was moreprominently affected in middle-aged rats than youngadult after contusion SCI, at all time points examinedout to 30 dpi. Middle-aged rats compared to young adultrats showed age-related delays in motor function withdelayed general functional recovery, as demonstrated bylower scores on the BBB test, and worsened ability tocomplete the footprint task. These results agree withprevious work in mouse and male and female rat modelsof aged SCI [19, 21, 40]. These previous studies demon-strate that 18-month-old female rats showed elevatedbase of support during walk both before and after injury[19], similar to our studies, while 12-month-old malerats consistently demonstrate impaired BBB function by7 dpi [40]. In contrast, 14-month-old male mice did notdemonstrate functional impairment before injury, al-though this may be due to the use of only the locomotorscore, similar to the BBB; however, a significant reduc-tion in locomotor score was noted after injury [21], as inour study.Interestingly, footprint analysis in middle-aged rats

after SCI are opposite of what was seen in naïve, show-ing decreased stride lengths and narrower toe spread at30 dpi versus increased stride length and wider toespread in naïve. Previous work in mouse models of SCIhave shown that both stride length and toe spreaddecrease significantly after injury [41–43] and have sug-gested that the decrease may be indicative of sensorymotor function deficits in the hind limbs due to injury.Thus, our findings of impaired recovery and alteredstepping pattern in footprint analysis may be associatedwith an age-related decrease in precise motor control inthe hind limbs.

ConclusionsOverall, the results of our study suggest that increasing ageleads to a pro-inflammatory environment, with alterationsto microglial activation, NOX2 enzyme component expres-sion, and an increase in oxidative stress in the cellular en-vironment that may contribute to worsened outcomes after

SCI. The initiating factor in these effects and the cause/ef-fect relationship between each remains unclear and is animportant consideration for the field moving forward. It isclear that there are intimate relationships between inflam-mation, oxidative damage, and NOX2, and the currentwork contributes to understanding that all three play a rolein motor function and post-injury recovery. However, morework is needed to understand which, if not all three, theessential component is. We currently hypothesize thatage-related changes to ROS production caused by in-creased microglial activation and NOX2 expressionare associated with an exaggerated chronic inflamma-tory response to injury.With the rise in SCI in the aging population, the results

of this study suggest that increased inflammatory responsemay be an important factor to be considered in thera-peutic interventions after injury and that age should beconsidered when developing a therapeutic treatment plan.Further, previous work has demonstrated that inhibitionof the NOX enzyme can decrease oxidative stress andinflammation. Due to the association of NOX2 with bothROS production and microglial activity, our findingssuggest a potential avenue for NOX2 inhibitory treatmentoptions to quell the exaggerated inflammatory responseand should be a topic of interest for future research.

AbbreviationsCNS: Central nervous system; DPI: Days post-injury; NOX: NADPH oxidase;ROS: Reactive oxygen species; SCI: Spinal cord injury

AcknowledgementsThe authors would like to acknowledge the editorial assistance provided byMs. I. von Leden. The opinions or assertions contained herein are the privateones of the author(s) and are not to be construed as official or reflected asthe views of the DoD or the USUHS.

FundingThis work was funded by the NINDS/NIH (Grant number 1R01NS073667-01A1) and a pilot grant from the Uniformed Services University. R. von Ledenwas supported by the NINDS/NIH (Grant number 1F31NS090737-01A1).

Availability of data and materialsAll data generated or analyzed during this study are included in thispublished article.

Authors’ contributionsRV conceived of the study; carried out and performed the statistical analysesfor all outcome measures, including the function, oxidative stress, geneexpression, and immunoassays; and drafted the manuscript. GK assisted inthe design and measurement of the functional outcomes andimmunoassays. KM participated in the design of the gene expression studiesand helped to draft the manuscript. KB participated in the design andcoordination of the study and helped to draft the manuscript. All authorsread and approved the final manuscript.

Ethics approvalAll animal procedures were approved by the Uniformed Services UniversityIACUC and complied fully with the principles set forth in the “Guide for theCare and Use of Laboratory Animals” prepared by the Committee on Careand Use of Laboratory Animals of the Institute of Laboratory Resources,National Research Council (DHEW pub. No. (NIH) 85-23, 2985).


Consent for publicationNot Applicable.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Received: 23 March 2017 Accepted: 4 August 2017

References1. NSCISC. National Spinal Cord Injury Statistical Center, facts and figures at a

glance. Birmingham: University of Alabama at Birmingham; 2016. p. 2016.2. Dumont RJ, Okonkwo DO, Verma S, Hurlbert RJ, Boulos PT, Ellegala DB, et al.

Acute spinal cord injury, part I: pathophysiologic mechanisms. ClinNeuropharmacol. 2001;24:254–64.

3. Ronsyn MW, Berneman ZN, Van Tendeloo VFI, Jorens PG, Ponsaerts P. Cancell therapy heal a spinal cord injury? Spinal Cord. 2008;46:532–9.

4. Jain NB, Ayers GD, Peterson EN, Harris MB, Morse L, O’Connor KC, et al.Traumatic spinal cord injury in the United States, 1993-2012. JAMA. 2015;313:2236–43.

5. Lu T, Pan Y, Kao S-Y, Li C, Kohane I, Chan J, et al. Gene regulation and DNAdamage in the ageing human brain. Nature. 2004;429:883–91.

6. Groah SL, Charlifue S, Tate D, Jensen MP, Molton IR, Forchheimer M, et al.Spinal cord injury and aging: challenges and recommendations for futureresearch. Am J Phys Med Rehabil Assoc Acad Physiatr. 2012;91:80–93.

7. Harman D. Free radical theory of aging. Mutat Res. 1992;275:257–66.8. Kregel KC, Zhang HJ. An integrated view of oxidative stress in aging: basic

mechanisms, functional effects, and pathological considerations. Am JPhysiol - Regul Integr Comp Physiol. 2007;292:R18–36.

9. Gao H-M, Zhou H, Hong J-S. NADPH oxidases: novel therapeutic targets forneurodegenerative diseases. Trends Pharmacol Sci. 2012;33:295–303.

10. Byrnes KR, Washington PM, Knoblach SM, Hoffman E, Faden AI. Delayedinflammatory mRNA and protein expression after spinal cord injury. JNeuroinflammation. 2011;8:130.

11. Loane DJ, Byrnes KR. Role of microglia in neurotrauma. Neurother J Am SocExp Neurother. 2010;7:366–77.

12. Sierra A, Gottfried-Blackmore AC, McEwen BS, Bulloch K. Microglia derived fromaging mice exhibit an altered inflammatory profile. Glia. 2007;55:412–24.

13. Salminen A, Ojala J, Kaarniranta K, Haapasalo A, Hiltunen M, Soininen H.Astrocytes in the aging brain express characteristics of senescence-associated secretory phenotype. Eur J Neurosci. 2011;34:3–11.

14. Norden DM, Muccigrosso MM, Godbout JP. Microglial priming andenhanced reactivity to secondary insult in aging, and traumatic CNS injury,and neurodegenerative disease. Neuropharmacology. 2015;96 Pt A:29–41.

15. Batchelor PE, Tan S, Wills TE, Porritt MJ, Howells DW. Comparison ofinflammation in the brain and spinal cord following mechanical injury. JNeurotrauma. 2008;25:1217–25.

16. Schnell L, Fearn S, Klassen H, Schwab ME, Perry VH. Acute inflammatoryresponses to mechanical lesions in the CNS: differences between brain andspinal cord. Eur J Neurosci. 1999;11:3648–58.

17. Zhang B, Gensel JC. Is neuroinflammation in the injured spinal corddifferent than in the brain? Examining intrinsic differences between thebrain and spinal cord. Exp Neurol. 2014;258:112–20.

18. Fenn AM, Hall JCE, Gensel JC, Popovich PG, Godbout JP. IL-4 signaling drives aunique arginase+/IL-1β+ microglia phenotype and recruits macrophages tothe inflammatory CNS: consequences of age-related deficits in IL-4Rα aftertraumatic spinal cord injury. J Neurosci. 2014;34:8904–17.

19. Hooshmand MJ, Galvan MD, Partida E, Anderson AJ. Characterization ofrecovery, repair, and inflammatory processes following contusion spinal cordinjury in old female rats: is age a limitation? Immun Ageing A. 2014;11:15.

20. Ritzel RM, Patel AR, Pan S, Crapser J, Hammond M, Jellison E, et al.Age- and location-related changes in microglial function. NeurobiolAging. 2015;36:2153–63.

21. Zhang B, Bailey WM, Braun KJ, Gensel JC. Age decreases macrophage IL-10expression: implications for functional recovery and tissue repair in spinalcord injury. Exp Neurol. 2015;273:83–91.

22. Zhang B, Bailey WM, McVicar AL, Gensel JC. Age increases reactive oxygenspecies production in macrophages and potentiates oxidative damage afterspinal cord injury. Neurobiol Aging. 2016;47:157–67.

23. Cooney SJ, Zhao Y, Byrnes KR. Characterization of the expression andinflammatory activity of NADPH oxidase after spinal cord injury. Free RadicRes. 2014;48:929–39.

24. Khayrullina G, Bermudez S, Byrnes KR. Inhibition of NOX2 reduces locomotorimpairment, inflammation, and oxidative stress after spinal cord injury. JNeuroinflammation. 2015;12:172.

25. Kumar A, Stoica BA, Sabirzhanov B, Burns MP, Faden AI, Loane DJ. Traumaticbrain injury in aged animals increases lesion size and chronically altersmicroglial/macrophage classical and alternative activation states. NeurobiolAging. 2013;34:1397–411.

26. Gwak YS, Hains BC, Johnson KM, Hulsebosch CE. Locomotor recovery andmechanical hyperalgesia following spinal cord injury depend on age attime of injury in rat. Neurosci Lett. 2004;362:232–5.

27. Kilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG. Improvingbioscience research reporting: the ARRIVE guidelines for reporting animalresearch. PLoS Biol. 2010;8:e1000412.

28. Cooney SJ, Bermudez-Sabogal SL, Byrnes KR. Cellular and temporalexpression of NADPH oxidase (NOX) isotypes after brain injury. JNeuroinflammation. 2013;10:155.

29. Basso DM, Beattie MS, Bresnahan JC. A sensitive and reliable locomotorrating scale for open field testing in rats. J Neurotrauma. 1995;12:1–21.

30. Kunkel-Bagden E, Dai HN, Bregman BS. Methods to assess the developmentand recovery of locomotor function after spinal cord injury in rats. ExpNeurol. 1993;119:153–64.

31. Loane DJ, Pocivavsek A, Moussa CE-H, Thompson R, Matsuoka Y, Faden AI,et al. Amyloid precursor protein secretases as therapeutic targets fortraumatic brain injury. Nat Med. 2009;15:377–9.

32. Moritz KE, Geeck K, Underly RG, Searles M, Smith JS. Post-operativeenvironmental enrichment improves spatial and motor deficits but may notameliorate anxiety- or depression-like symptoms in rats following traumaticbrain injury. Restor Neurol Neurosci. 2014;32:701–16.

33. Donnelly DJ, Gensel JC, Ankeny DP, van Rooijen N, Popovich PG. Anefficient and reproducible method for quantifying macrophages in differentexperimental models of central nervous system pathology. J NeurosciMethods. 2009;181:36–44.

34. Selwyn R, Hockenbury N, Jaiswal S, Mathur S, Armstrong RC, Byrnes KR. Mildtraumatic brain injury results in depressed cerebral glucose uptake: an(18)FDG PET study. J Neurotrauma. 2013;30:1943–53.

35. Morganti JM, Riparip L-K, Chou A, Liu S, Gupta N, Rosi S. Age exacerbatesthe CCR2/5-mediated neuroinflammatory response to traumatic brain injury.J Neuroinflammation. 2016;13:80.

36. Jin X, Ishii H, Bai Z, Itokazu T, Yamashita T. Temporal changes in cell markerexpression and cellular infiltration in a controlled cortical impact model inadult male C57BL/6 mice. PLoS One. 2012;7:e41892.

37. Fouad K, Hurd C, Magnuson DSK. Functional testing in animal models ofspinal cord injury: not as straight forward as one would think. Front IntegrNeurosci. 2013;7:85.

38. Loy DN, Talbott JF, Onifer SM, Mills MD, Burke DA, Dennison JB, et al. Bothdorsal and ventral spinal cord pathways contribute to overgroundlocomotion in the adult rat. Exp Neurol. 2002;177:575–80.

39. Schucht P, Raineteau O, Schwab ME, Fouad K. Anatomical correlates oflocomotor recovery following dorsal and ventral lesions of the rat spinalcord. Exp Neurol. 2002;176:143–53.

40. Gwak YS, Hains BC, Johnson KM, Hulsebosch CE. Effect of age at timeof spinal cord injury on behavioral outcomes in rat. J Neurotrauma.2004;21:983–93.

41. Blakeman KH, Hao J-X, Xu X-J, Jacoby AS, Shine J, Crawley JN, et al.Hyperalgesia and increased neuropathic pain-like response in mice lackinggalanin receptor 1 receptors. Neuroscience. 2003;117:221–7.

42. Hygge-Blakeman K, Brumovsky P, Hao J-X, Xu X-J, Hökfelt T, Crawley JN,et al. Galanin over-expression decreases the development of neuropathicpain-like behaviors in mice after partial sciatic nerve injury. Brain Res. 2004;1025:152–8.

43. Beare JE, Morehouse JR, DeVries WH, Enzmann GU, Burke DA, MagnusonDSK, et al. Gait analysis in normal and spinal contused mice using theTreadScan system. J Neurotrauma. 2009;26:2045–56.


Documents

Age exacerbates microglial activation, oxidative stress